cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Search instead for 
Did you mean: 
novinosrin
‎12-24-2025
  • 7,347 Posts
  • 1,429 Likes Given
  • 1,421 Solutions
  • 3,055 Likes Received

Re: PROC SQL SELECT DISTINCT

by Tourmaline | Level 20 in SAS Programming
‎08-08-2017 03:29 PM
‎08-08-2017 03:29 PM
remove the parenthesis around distinct. It's not a function ... View more

Re: Trying to read in a two variable file using INFILE but can't get i...

by Tourmaline | Level 20 in SAS Programming
‎08-08-2017 02:14 PM
1 Like
‎08-08-2017 02:14 PM
1 Like
Ok this code is working for me. Thank you! Few follow up questions -   1) Why do you need the format statement... (below in bold/underline)? Shouldn't defining the data type as mmddyy10. in the input statement take care of the format?  The format in the input statement is actually an informat that reads and converts non standard data like your character date to a SAS date value which is merely a number. This number has to be rerepresented as a character date for which you need a format statement. Always remember: Informat is to read and format is to write. They look the same yet they are different   2) Why did you use expandtabs instead of delimiter='   '? I was lazy at first to have noticed your data delimited to blank at face value. But in reality when i actually moved my cursor I learned that it was actually tab delimited perhaps a copy paste from excel. If it's tab demilited you can't quite use dlm=' ' instead you need expandtabs or something dlm='2008'x it's blank and tab delimited--> @art297 the genius taught me this a couple of years ago   3) What is truncover? Basically reads variable length records and doesn't jump to the next record and start reading  new values. Chill, in your data i don't see missing values, so this shouldn't affect the execution   HTH, Naveen Srinivasan   data SAData; infile 'D:\qsilver\Sas Data\NSAPur.txt' expandtabs truncover; input datefield : mmddyy10. purindex ; format datefield mmddyy10.; run; ... View more

Re: Trying to read in a two variable file using INFILE but can't get i...

by Tourmaline | Level 20 in SAS Programming
‎08-08-2017 01:49 PM
‎08-08-2017 01:49 PM
data SAData; infile 'C:\Users\NSRINIV2\Desktop\NSAPurSample.txt' expandtabs truncover; input datefield : mmddyy10. purindex ; format datefield mmddyy10.; run;   Hello mate!, Your file data is tab delimited in the first place. Kindly check and run the above   here is my log: NOTE: The infile 'C:\Users\NSRINIV2\Desktop\NSAPurSample.txt' is: Filename=C:\Users\NSRINIV2\Desktop\NSAPurSample.txt, RECFM=V,LRECL=32767,File Size (bytes)=110, Last Modified=08Aug2017:12:35:10, Create Time=08Aug2017:12:35:10 NOTE: 7 records were read from the infile 'C:\Users\NSRINIV2\Desktop\NSAPurSample.txt'. The minimum record length was 13. The maximum record length was 15. NOTE: The data set WORK.SADATA has 7 observations and 2 variables. NOTE: DATA statement used (Total process time): real time 0.01 seconds cpu time 0.01 seconds   ... View more

Re: Trying to read in a two variable file using INFILE but can't get i...

by Tourmaline | Level 20 in SAS Programming
‎08-08-2017 01:29 PM
‎08-08-2017 01:29 PM
should work:   data SAData; infile 'D:\qsilver\Sas Data\NSAPur.txt' delimiter=' '; input datefield :mmddyy10. purindex format datefield mmddyy10.; ; run; ... View more

Re: INTCK to compute minutes between dates

by Tourmaline | Level 20 in SAS Programming
‎08-08-2017 12:50 PM
‎08-08-2017 12:50 PM
Here is my test:   10 data want; 11 interval=intck('min',"21JUL2017: 9:06"dt,"23JUL2017 :10:28"dt); 12 put interval=; 13 run; interval=2962 NOTE: The data set WORK.WANT has 1 observations and 1 variables. NOTE: DATA statement used (Total process time): real time 0.04 seconds cpu time 0.01 seconds ... View more

Re: Count value across variables

by Tourmaline | Level 20 in SAS Programming
‎08-06-2017 05:24 PM
‎08-06-2017 05:24 PM
data dt1; input a1-a5; cards; 1  2    3 4 5 4 4  4   4 2 3 4 4    4    4    3    4 4 4    4.1  4.2  4    4    4 4 4 4 4 4 ; run;     data want1; set dt1; array num _numeric_; count=0; if std(of num(*))=0 then flag='N'; else do; flag='Y'; call sortn(of num(*)); do _n_=1 to dim(num)-1; if std(num(_n_),num(_n_+1))= 0 then continue; else count+1; end; end;  run;   /*Without callsortn- slightly different count*/ data want1; set dt1; array num _numeric_; count=0; if std(of num(*))=0 then flag='N'; else do; flag='Y'; do _n_=1 to dim(num)-1; if std(num(_n_),num(_n_+1))= 0 then continue; else count+1; end; end;  run;   ... View more

Re: change numeric values to sas time format

by Tourmaline | Level 20 in SAS Programming
‎08-04-2017 05:39 PM
1 Like
‎08-04-2017 05:39 PM
1 Like
Mine isn't smart, but fun though-   data have; input time; datalines; 120 2130 10 1128 23 50 ; data want; set have; new_time=cats(reverse(substr(reverse(time),3)),':',substr(time,length(time)-1)); if length(new_time)=3 then new_time=cats('00',new_time); run; ... View more

Re: Counting three consecutive month with no transaction

by Tourmaline | Level 20 in SAS Programming
‎08-04-2017 01:03 AM
‎08-04-2017 01:03 AM
Some more fun-   data account; input account $ transactionamount    month $; datalines; FORD   100                           JAN FORD   100                           FEB FORD   100                           MAR FORD   .                               APR FORD   .                               MAY FORD   .                              JUN FORD   100                           JUL FORD   100                           AUG FORD   100                           SEP FORD   100                           OCT FORD   100                           NOV FORD   100                           DEC TOY      100                           JAN TOY   100                           FEB TOY   100                           MAR TOY   100                           APR TOY   100                           MAY TOY   100                          JUN TOY   100                           JUL TOY   100                           AUG TOY   100                           SEP TOY   100                           OCT TOY   100                           NOV TOY   100                           DEC ;   data want;  declare hash myhash();     myhash.definekey('account','transactionamount');     myhash.definedata('remark');     myhash.definedone(); do until(last.account); do _n_=1 by 1 until(mod(_n_,3)=0);   set account; by account notsorted; if _n_=1 then temp=0; temp+transactionamount; if _n_=3 and temp=0 then do; remark='broken'; myhash.add(); end; else call missing(remark); end; end; do until(last.account); set account; by account notsorted; if myhash.find() ne 0 then call missing(remark); output; end; drop temp; run;   Regards, Naveen Srinivasan ... View more

Re: Counting three consecutive month with no transaction

by Tourmaline | Level 20 in SAS Programming
‎08-04-2017 12:08 AM
‎08-04-2017 12:08 AM
data account; input account $ transactionamount    month $; datalines; FORD   100                           JAN FORD   100                           FEB FORD   100                           MAR FORD   .                               APR FORD   .                               MAY FORD   .                              JUN FORD   100                           JUL FORD   100                           AUG FORD   100                           SEP FORD   100                           OCT FORD   100                           NOV FORD   100                           DEC TOY      100                           JAN TOY   100                           FEB TOY   100                           MAR TOY   100                           APR TOY   100                           MAY TOY   100                          JUN TOY   100                           JUL TOY   100                           AUG TOY   100                           SEP TOY   100                           OCT TOY   100                           NOV TOY   100                           DEC ;       data want; do until(last.account); do _n_=1 by 1 until(mod(_n_,3)=0); if _n_=1 then temp=0; set account; by account notsorted; temp+transactionamount; if _n_=3 and temp=0 then remark='broken'; else call missing(remark); output; end; end; drop temp; run;     ... View more

Re: Counting the number of rows that contain one or missing values

by Tourmaline | Level 20 in SAS Programming
‎08-03-2017 09:39 PM
‎08-03-2017 09:39 PM
Something like this pehaps:   data want; set have end=last;  if cmiss(of _char_) or nmiss(of _numeric_)>=1 then count+1; if last; keep count; run; ... View more

Re: error in converting text date column d/mm/yyyy to SAS date column ...

by Tourmaline | Level 20 in SAS Programming
‎08-03-2017 09:17 PM
‎08-03-2017 09:17 PM
data have; input period : ddmmyy10. (item_code patient_cat prescriptions) ($); format period yymmd.; datalines; 9/1/2013 03387G C0 4941 9/1/2013 03387G C1 5113 9/1/2013 03387G G1 644 9/1/2013 03387G G2 3097 9/1/2013 03387G R0 233 9/1/2013 03387G R1 186 9/1/2013 03390K C0 10 9/1/2013 03390K C1 18 9/1/2013 03390K G1 1 9/1/2013 03390K G2 62 9/1/2013 03391L C1 3 9/1/2013 03391L G2 4 9/1/2013 03393N C0 8 9/1/2013 03393N C1 296 9/1/2013 03393N C1 5 9/1/2013 03393N G1 1 9/1/2013 03393N G2 555 9/1/2013 03393N R0 1 9/1/2013 03393N R1 1 10/1/2013 03387G C0 112 10/1/2013 03387G C1 4 10/1/2013 03387G G1 7 10/1/2013 03387G G2 3000 10/1/2013 03387G R0 23 10/1/2013 03387G R1 0 10/1/2013 03390K C0 325556 10/1/2013 03390K C1 6 10/1/2013 03390K G1 3 10/1/2013 03390K G2 2 10/1/2013 03391L C1 44566 10/1/2013 03391L G2 323 10/1/2013 03393N C0 21 10/1/2013 03393N C1 1 10/1/2013 03393N C1 1 10/1/2013 03393N G1 9 10/1/2013 03393N G2 9 10/1/2013 03393N R0 7 10/1/2013 03393N R1 4 ; ... View more

Re: Counting three consecutive month with no transaction

by Tourmaline | Level 20 in SAS Programming
‎08-03-2017 08:12 PM
‎08-03-2017 08:12 PM
Please post a sample of your input data and a small sample of output data( exactly how you want). Transaction data is a piece of cake. Once I have your req in figures, I will give you the code. It also helps others too. Thanks! ... View more

Re: How to ensure two datasets to contain the same observations with ...

by Tourmaline | Level 20 in SAS Programming
‎08-03-2017 05:16 PM
‎08-03-2017 05:16 PM
If I understand you, is this what you are after, a comparison:   data WORK.TEST1;   infile datalines dsd truncover;   input CUSIP:$8. fe:32. yymm_a:YYMMN6.;   format yymm_a YYMMN6.;   label CUSIP="CUSIP/SEDOL" fe="Mean Estimate"; datalines4; 00036020,0.06,199201 00036020,0.06,199201 00036020,0.06,199201 00036020,0.07,199201 00036110,0.91,199705 00036110,1.24,199705 00036110,1.47,199705 00036110,1.73,199705 00036110,1.55,199705 00036110,1.61,199705 00036110,1.54,199705 00036110,1.25,199705 00036110,1.19,199705 00036110,1.65,199705 00036110,1.71,199705 00036110,1.24,199705 00036110,1.46,199705 00036110,1.47,199810 00036110,1.04,199810 00036110,1.73,199810 ;;;; data WORK.TEST2;   infile datalines dsd truncover;   input CUSIP:$8. fe:32. yymm_a:YYMMN6.;   format yymm_a YYMMN6.;   label CUSIP="CUSIP/SEDOL" fe="Mean Estimate"; datalines4; 00036020,0.09,199801 00036020,0.05,199801 00036020,0.03,199801 00036020,0.03,199801 00036020,0.07,199801 00036020,0.03,199801 00036020,0.06,199801 00036020,0.03,199801 00036020,0.04,199801 00036110,0.88,199705 00036110,0.46,199705 00036110,0.57,199705 00036110,0.47,199705 00036110,0.91,199705 00036110,0.93,199705 00036110,0.41,199705 00036110,0.65,199705 00036110,0.51,199705 00036110,0.47,199705 00036110,0.88,199810 ;;;;     proc sort data=test1 nodupkey; by cusip yymm_a; run; proc sort data=test2 nodupkey; by cusip yymm_a; run; data exist_only_test1 exist_only_test2 exist_test1_test2 ; merge test1(in=a) test2(in=b ); by cusip yymm_a; if a and b then output exist_test1_test2; else if a and not b then output exist_only_test1; else if b and not a then output exist_only_test2; run; ... View more

Re: Excel Export

by Tourmaline | Level 20 in SAS Programming
‎08-03-2017 03:38 PM
‎08-03-2017 03:38 PM
@PaigeMiller i love that. Cheers! lol ... View more

Re: base sas

by Tourmaline | Level 20 in SAS Programming
‎08-03-2017 03:36 PM
‎08-03-2017 03:36 PM
Where are the groups to use by variables in your sample. Your id values are all unique. The marks has ties. The point of using first and last variables would not make sense your requirement. To use first and last, you need repeats on by values. For example, 1. a transaction dataset 2. Patients multiple visits to hospitals 3. Repeat customers in a store like a loyatly program   If you understand the business, the code will come to your mind automatically. Anyways, let us the know the correct req, you will get a neat solution. Also, you should let us know how to handle ties in ranks, i.e b and f in your example, both having scored 33 ... View more
Contact Me
Online Status
Offline
Date Last Visited
‎12-24-2025 04:47 PM
Group Hubs for novinosrin
View All
Likes given to
Likes from
Quick links